Science - 31 January 2020

SCIENCE sciencemag.org 31 JANUARY 2020 • VOL 367 ISSUE 6477 515

GRAPHIC: X. LIU/S

CIENCE

for private-sector natural capital account-

ing, for example, with the Natural Capital

Protocol. This allows businesses to better

understand their impacts and dependency

on natural capital.

CHALLENGES AND OPPORTUNITIES

Ecosystem accounts published to date vary

in scope and level of (spatial) detail. This re-

flects differences in budget, technical capac-

ity, and data between countries, with notable

constraints in developing countries. Compil-

ing the full suite of ecosystem accounts re-

quires substantial data and use of multiple

biophysical models (see SM, section 2). The

Netherlands ecosystem accounts provide 90

policy-relevant indicators, derived from vari-

ous datasets and models. They indicate, for

example, how small landscape elements such

as hedgerows contribute to crop production

by maintaining pollinator populations.

Institutional challenges occur, for instance,

in relation to integrating data from different

agencies: Data may be in incompatible for-

mats, or there may be a reluctance to share

it. Furthermore, the SEEA has a different

measurement approach compared with some

existing reporting systems (for example,

carbon reporting under the UN Framework

Convention on Climate Change). Collabora-

tion between government institutes holding

different datasets facilitates enhanced data

integration and greater commonality in ter-

minology and definitions ( 13 ).

Some technical challenges remain. For

instance, the diversity of ecosystems makes

selection of ecosystem condition and bio-

diversity indicators challenging. In South

African and Australian accounts, ecosystem

condition has been defined in relation to pre-

European settlement conditions, which is not

useful for western Europe with its long eco-

system-use history. Challenges also exist in

valuing nonmarket ecosystem services such

as water regulation and air filtration.

The EEA’s inherent limitations should be

considered when the accounts are used in

policy-making. The EEA accounts produced

to date do not include indicators for ecosys-

tem resilience or consider probabilities of

sudden future collapses of overexploited eco-

systems ( 15 ). Furthermore, exchange prices

of ecosystem services reflect current pricing

mechanisms and market conditions. Given

that the SEEA does not record the welfare

generated by using natural capital ( 2 ), it is

imperative that monetary values in the EEA

are not interpreted as representing “the value

of nature” (see SM, section 3).

Several ongoing efforts address remain-

ing challenges to the global implementation

of the EEA. The UNSC is working with sci-

entists and statisticians toward establishing

a statistical standard for the EEA by 2021.

Working groups have been established to

address remaining technical issues, includ-

ing defining metrics expressing ecosystem

condition, biodiversity, and the capacity of

ecosystems to supply services and valuing

nonmarket ecosystem services (for instance,

on the basis of simulated exchange values).

Through various scientific efforts connected

to the EEA, such as the Earth Observation for

Ecosystem Accounting initiative of the Group

on Earth Observations, tests are being done

to examine how ecosystem extent, condi-

tion, and regulating services can be modeled

across large countries or even continents at

high resolution using data from remote-

sensing and global datasets. Increasingly,

machine-learning techniques are used, for

instance, to assess the impacts of ecosystem

changes on hydrological cycles and the avail-

ability of water for people. Social media posts

can be used to analyze recreation in ecosys-

tems, for example. The EEA is making large

datasets available to a variety of users, and

global, high-resolution modeling of critical

ecosystem characteristics and services will

facilitate easier uptake of the EEA in develop-

ing countries.

The EEA allows consistent (over time and

between countries), comprehensive, and

high-resolution analysis and reporting on

ecosystems and their use. The EEA accounts

do not capture all connections between peo-

ple and nature and have limited capacity to

consider ecosystem complexities such as

thresholds and feedbacks. These caveats need

to be clearly articulated when EEA accounts

are published. Nonetheless, the EEA consid-

erably enhances the scope and accuracy of in-

formation available in support of ecosystem

management, facilitating better management

of global natural capital. j

REFERENCES AND NOTES

1. J. Stiglitz, J. Fitoussi, M. Duran, Beyond GDP: Measuring
   What Counts for Economic and Social Performance (OECD
   Publishing, 2018).


2. D. W. Jorgenson, J. Econ. Lit. 56 , 867 (2018).


3. UN et al., The System of National Accounts, 2008 (UN,
   2009).


4. UN et al., “System of environmental-economic account-
   ing 2012: Central framework” (UN, 2014).


5. UN et al., “System of environmental-economic account-
   ing: Experimental ecosystem accounting” (UN, 2014).


6. UN, “Technical Recommendations in support of the SEEA
   Experimental Ecosystem Accounting” (UN, 2019).


7. ONS (Office for National Statistics), “UK natural capital
   accounts: 2019” (ONS, 2019).


8. Horlings et al., “Experimental monetary valuation of eco-
   system services and assets in the Netherlands” (Report,
   Statistics Netherlands and Wageningen University and
   Research, 2019).


9. H. Keith et al., Nat. Ecol. Evol. 1 , 1683 (2017).


10. P. Campos et al., Ecol. Econ. 157 , 218 (2019).


11. J. L. Nel, A. Driver, “National river ecosystem accounts
    for South Africa” (South African National Biodiversity
    Institute, Pretoria, 2015).


12. S. Vallecillo et al., “Ecosystem services accounting—Part
    II Pilot accounts for crop and timber provision, global
    climate regulation and flood control” (JRC technical
    reports, Publications Office of the European Union, 2019).


13. G. Bright et al., Oxf. Rev. Econ. Policy 35 , 88 (2019).


14. Government of the Netherlands, “National Climate
    Agreement” (2019).
    1 5. C. Fo l ke et al., Ecol. Soc. 21 , 41 (2016).
    ACKNOWLEDGMENTS
    The views expressed in this article are those of the authors
    and do not necessarily reflect the policies of the UN or the
    European Union.
    SUPPLEMENTARY MATERIALS
    science.sciencemag.org/content/367/6477/514/suppl/DC1
    10.1126/science.aaz8901

National National and subnational Subnational

Type of ecosystem accounts compiled

Canada

Mexico Guatemala

United States

Costa Rica

Colombia

Peru

Australia

Japan

China

India

Indonesia

Philippines

South Africa Mauritius

Madagascar

Liberia

Rwanda

Uganda

Croatia

Finland

United Kingdom Netherlands Norway

Denmark

Spain

Italy

Countries that have compiled SEEA EEA accounts

Some countries have published all accounts that they have compiled, and others have published only some.

China, Japan, and the United States have compiled accounts but not published them (see supplementary

materials, section 1). The scope and resolution of the accounts vary between countries. The figure presents

a snapshot—countries continue to compile and publish accounts. SEEA, System of Environmental-Economic

Accounting; EEA, Experimental Ecosys tem Accounting.

Published by AAAS